The Unscented Transform: Theory and Bayesian Applications

The Unscented Transform: theory, extensions and practical application to Bayesian inference

We are delighted to have John Whitamore back at the Bayesian Mixer. Please register here: https://www.tickettailor.com/events/bayesianmixer/2154878

Abstract: How can we perform Bayesian inference without resorting to slow random sampling methods? Isn't there a nice, pragmatic way to use the geometry of a problem to use only a very small number of samples, placed deterministically? Isn't there a good, straightforward way to implement Bayesian models in practical settings?

This talk explains the real-world problems that can be solved by the Unscented Transform. It begins with an intuitive introduction to Bayesian methods, discusses the Unscented Transform and connects the ideas to geometry, finite elements analysis and deep learning.

Bio: John Whitamore is a Senior Data Scientist at Simply Business, a leading UK InsureTech firm. He has previously served as Head of Data Science for a leading food retailer, as Head of Trading and Systems for an investment management company and as Global Head of Convertible Bond Trading for a European investment bank.